Transcripts of ceruloplasmin but not hepcidin, both major

HAL Id: hal-00501561https://hal.archives-ouvertes.fr/hal-00501561

Submitted on 12 Jul 2010

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Transcripts of ceruloplasmin but not hepcidin, bothmajor iron metabolism genes, exhibit a decreasing

pattern along portocentral axis of mouse liverMarie-Bérengère Troadec, Alain Fautrel, Bernard Drénou, Patricia Leroyer,Emilie Camberlein, Bruno Turlin, André Guillouzo, Pierre Brissot, Olivier

Loréal

To cite this version:Marie-Bérengère Troadec, Alain Fautrel, Bernard Drénou, Patricia Leroyer, Emilie Camberlein, et al..Transcripts of ceruloplasmin but not hepcidin, both major iron metabolism genes, exhibit a decreasingpattern along portocentral axis of mouse liver. Biochimica et Biophysica Acta - Molecular Basis ofDisease, Elsevier, 2008, 1782 (4), pp.239. �10.1016/j.bbadis.2007.12.009�. �hal-00501561�

https://hal.archives-ouvertes.fr/hal-00501561

https://hal.archives-ouvertes.fr

��

Transcripts of ceruloplasmin but not hepcidin, both major iron metabolismgenes, exhibit a decreasing pattern along portocentral axis of mouse liver

Marie-Berengere Troadec, Alain Fautrel, Bernard Drenou, Patricia Leroyer,Emilie Camberlein, Bruno Turlin, Andre Guillouzo, Pierre Brissot, OlivierLoreal

PII: S0925-4439(07)00239-6DOI: doi: 10.1016/j.bbadis.2007.12.009Reference: BBADIS 62776

To appear in: BBA - Molecular Basis of Disease

Received date: 24 July 2007Revised date: 23 November 2007Accepted date: 18 December 2007

Please cite this article as: Marie-Berengere Troadec, Alain Fautrel, Bernard Drenou, Pa-tricia Leroyer, Emilie Camberlein, Bruno Turlin, Andre Guillouzo, Pierre Brissot, OlivierLoreal, Transcripts of ceruloplasmin but not hepcidin, both major iron metabolism genes,exhibit a decreasing pattern along portocentral axis of mouse liver, BBA - Molecular Basisof Disease (2008), doi: 10.1016/j.bbadis.2007.12.009

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

http://dx.doi.org/10.1016/j.bbadis.2007.12.009

http://dx.doi.org/10.1016/j.bbadis.2007.12.009

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPTTroadec MB et al,

Liver zonation, ploidy and iron genes Page 1

TITLE PAGE

Title

TRANSCRIPTS OF CERULOPLASMIN BUT NOT HEPCIDIN, BOTH MAJOR

IRON METABOLISM GENES, EXHIBIT A DECREASING PATTERN ALONG

PORTOCENTRAL AXIS OF MOUSE LIVER.

Author Names

Marie-Bérengère Troadec 1, Alain Fautrel 2,3, Bernard Drénou 4, Patricia Leroyer 1, Emilie

Camberlein 1 , Bruno Turlin 1,3,5, André Guillouzo 2, Pierre Brissot 1, 6 and Olivier Loréal 1

Affiliations 1INSERM U522 ; University of Rennes 1 ; IFR 140; 2 INSERM U620 ; University of Rennes

1 ; IFR 140; 3 IFR 140 Core HistoPathology Platform ; 4 Haematology Department,

Mulhouse Hospital, 5 Department of Anatomopathology and 6Liver Disease Unit, University

Hospital Pontchaillou, 35033 Rennes, France.

Short title : iron metabolism, liver zonation and ploidy

Keywords :Iron metabolism; liver zonation; ploidy; ceruloplasmin; hepcidin; gene

expression; laser microdissection, cytometry; mouse.

Corresponding author

Dr Olivier LOREAL; INSERM U522, Hospital Pontchaillou, 35033 Rennes, France.

Phone :33.2.99.54.37.37 ; Fax :33.2.99.54.01.37 e-mail : [email protected]

Dr Marie-Bérengère TROADEC; INSERM U522, Hospital Pontchaillou, 35033 Rennes,

France. Phone :33.2.99.54.37.37 ; Fax :33.2.99.54.01.37 e-mail : marie-

[email protected]

ACC

EPTE

D M

ANU

SCR

IPT



SUMMARY

Background/Aims: During iron overload of dietary origin, iron accumulates predominantly

in periportal hepatocytes. A gradient in the basal and normal transcriptional control of genes

involved in iron-metabolism along the portocentral axis of liver lobules could explain this

feature. Therefore, we aimed at characterizing, by quantitative RT-PCR, the expression of

iron-metabolism genes in adult C57BL/6 mouse hepatocytes regarding lobular localisation,

with special emphasis to cell ploidy, considering its possible relationship with lobular

zonation. Methods: We used two methods to analyse separately periportal and perivenous

liver cells: 1) a selective liver zonal destruction by digitonin prior to a classical collagenase

dissociation, and 2) laser capture microdissection. We also developed a method to separate

viable 4N and 8N polyploidy hepatocytes by flow cytometer. Results: Transcripts of

ceruloplasmin, involved in iron efflux, were overexpressed in periportal areas and the result

was confirmed by in situ hybridization study. By contrast, hepcidin 1, hemojuvelin,

ferroportin, transferrin receptor 2, hfe and L-ferritin mRNAs were not differentially expressed

according to either lobular zonation or polyploidisation level. Conclusions: At variance with

glutamine or urea metabolism, iron metabolism is not featured by a metabolic zonation lying

only on a basal transcriptional control. The preferential periportal expression of ceruloplasmin

raises the issue of its special role in iron overload disorders involving a defect in cellular iron

export.

Abstract word count: 216

List of Abbreviations

LDH: lactate dehydrogenase; SSC: Side Scatter Channel; FSC: Forward Scatter Channel; PI:

Propidium Iodide; 2N: diploid cells; 4N: tetraploid cells; 8N: octoploid cells; LCM: Laser

Capture Microdissection.

ACC

EPTE

D M

ANU

SCR

IPT



INTRODUCTION

Hepatocytes are heterogeneous for metabolic functions (review in [1]) and ploidy (total cell

DNA content) [2, 3] within hepatic lobules.

Periportal hepatocytes are mainly involved in ureagenesis, bile formation and glycogenesis.

Perivenous (centrilobular) hepatocytes are preferentially implicated in glycolysis, lipogenesis,

glutamine synthesis and xenobiotic metabolism (review in [1]). The liver metabolic zonation

can be partially related to a static transcriptional zonation implicating transcriptional factors

such as recently described in the Wnt/beta-catenin pathway [4], or physiological parameters

including oxygen tension, blood stream, circulating factors, paracrine effects between cells or

cell-cell contacts (review in [1]).

Cell ploidy is defined as the total cellular DNA content. The normal DNA content for

eukaryotic cells is 2N (diploid cell). Cells with more than 2N chromosomes are called

polyploid. DNA can be distributed in one nucleus (mononuclear) or 2 nuclei (binuclear).

Hepatocytes can therefore be diploid (2N), tetraploid (4N) with one nucleus or tetraploid with

2 diploid nuclei, octaploid (8N). Furthermore, the liver exhibits a peculiar distribution of

hepatocyte ploidy within the lobules, and expresses biological differences between diploid

and polyploid hepatocytes [5, 6]. Previous results suggested, by indirect evaluations [5, 6], a

particular distribution of hepatocyte ploidy within the lobules. Indeed, diploid-enriched

fractions and polyploid cells showed phenotypic markers from periportal areas and perivenous

areas respectively [5].

During certain human iron overload diseases, such as HFE haemochromatosis, iron

accumulates in hepatocytes following a portocentral decreasing gradient [7]. This

heterogeneous distribution of iron is also observed in iron overload models of : i) genetic

ACC

EPTE

D M

ANU

SCR

IPT



origin, such in mice knock-out for Hfe gene (Hfe -/- ) [8], ii) or iron-rich diet origin, such as

carbonyl-iron supplemented mice [9]. Periportal hepatocytes are the first cells to receive both

transferrin and non-transferrin bound iron [10] from blood flow and are therefore iron-

overloaded prior to centrilobular hepatocytes. However, unexpectedly, mice knock-out for the

hepcidin gene (hepc1-/-) mice show a perivenous iron accumulation [11] suggesting that other

types of control could be implicated in addition to the blood flow.

In this study, we addressed the question of an iron metabolism zonation lying on a static

mRNA expression level, on the model of glutamine or urea metabolism [4].

Distribution of iron-metabolism gene expression within liver lobules is partial, and mainly

documented in rats [12, 13],. However, except for hepcidin and hemojuvelin [14] no data on

the distribution of most major iron-metabolism genes is yet available in mouse, despite its

recognition as a model for studying the pathophysiology of iron metabolism.

Our aim was to characterize the hepatic expression of iron-metabolism genes in adult mouse

regarding: i) hepatocyte localisation within the liver lobules and ii) hepatocyte ploidy status.

We focused our study on genes implicated in: i) the local or systemic control of iron

homeostasis such as hepcidin 1 (Hepc1 also named Hamp1) [15, 16], Hfe [17], hemojuvelin

(Hjv) [18] and transferrin receptor 2 (Tfr2) [19, 20], ii) iron uptake : transferrin receptor 1

(Tfr1) [21], iii) iron storage : L-ferritin [22, 23], and iv) iron efflux : ferroportin [24-26], and

ceruloplasmin [27]. Periportal and perivenous hepatocytes were separated either by selective

areas destruction by digitonin prior to liver dissociation or by laser capture microdissection.

Moreover, in order to isolate viable hepatocytes on the basis of DNA content, we developed a

on flow cytometric method allowing further mRNA level quantification.

ACC

EPTE

D M

ANU

SCR

IPT



We found that ceruloplasmin showed a decreasing portocentral transcriptional gradient along

the lobules. By contrast, no mRNA level gradient was found according to hepatocyte ploidy

or along the portocentral axis of liver lobules for other iron-related genes, including hepcidin

1, hfe, hemojuvelin, transferrin receptor 2, L-ferritin and ferroportin. Our results suggest that

the zonal iron accumulation observed during hepatic iron overload cannot be explained by a

major static zonation of iron-metabolism transcriptional regulation in C57BL/6 mice, and

suggest to take into account the role of the periportal expression of ceruloplasmin in iron

overload conditions involving a defect in cellular iron efflux.

ACC

EPTE

D M

ANU

SCR

IPT



MATERIALS AND METHODS

Animals

Adult 20-week old C57BL/6 male mice from CERJ (Le Genest St Isle, France) were used.

They were maintained under standard conditions of temperature, atmosphere and light, and

experimental procedures were performed in agreement with French law and regulations

(Agreement B-35-238-10). They had free access to tap water and standard AO3 diet (UAR,

France).

Selective isolation of perivenous and periportal hepatocytes by digitonin-collagenase

perfusion

After anaesthesia, perivenous and periportal hepatocytes were prepared by the digitonin-

collagenase perfusion method [28, 29] adapted to mouse. Hepatic veins were first ligatured.

To obtain perivenous hepatocytes, the liver was first washed 3 min (10 mL/min) in HEPES

buffer and then short-term perfused with 7 mM digitonin for a few seconds at 2.5 mL/min

through the portal vein. The liver was then washed by an anterograde wash flow 10 min (10

mL/min) in calcium-free HEPES buffer through the inferior vena cava, followed by 8 min (10

ml/min) of enzymatic dissociation in HEPES buffer (0.025% collagenase, 0.075%CaCl2). To

obtain periportal hepatocytes, destruction of perivenous areas was achieved by perfusion

through the inferior vena cava and dissociation through the portal vein. Cells were then

filtrated on nylon 60µ with Leibowitz medium (Invitrogen) and settled 20 min in order to

enrich the pellet in hepatocytes. Cells were washed twice in HEPES (700 rpm, 1min) and

once in MEM:M119 (3v:1v; Invitrogen) in order to eliminate dead cells, and thereafter

immediately frozen at -80°C until RNA extraction.

ACC

EPTE

D M

ANU

SCR

IPT



Selective isolation of perivenous and periportal hepatic cells by Laser Capture

Microscopy (LCM)

After anaesthesia, livers were removed, frozen in isopentane then liquid nitrogen, and stored

at -80°C. Ten µm thick frozen sections were cut on a cryostat (Leica, Milton Keynes, UK),

mounted onto uncoated glass slides, fixed at -20°C in 70% ethanol for 1 min, then stained

with histogene (Arcturus Engineering, Mountain View, California, USA) for 5s at room

temperature, washed briefly in 70% ethanol and sequentially dehydrated in 100% ethanol and

xylene. The sections were then microdissected using a Veritas Laser Capture Microdissection

system (LCM) (Arcturus). The settings of the InfraRed laser were: spot diameter 20 µm,

pulse duration 3500 ms and power 90 mW. On the same section, 1 mm2 of perivenous and

centrilobular areas were microdissected with a separated "cap". All areas were selected and

collected in less than 30 min after the slide preparation. RNA isolation was performed using

the PicoPure RNA isolated kit (Arcturus). RNAs were quality-checked (Ribosomal Integrated

Number (Agilent) at 7.5).

In situ Hybridization

Digoxigenin-labeled riboprobe preparation. Two-hundred/Four-hundred base pairs length

gene fragment of mouse ceruloplasmin and ferroportin were cloned into pGEM-T or pGEM-

Teasy vectors (Promega, Madison, WI) with T7 and SP6 promoters flanking either side.

Sequence and orientation were confirmed by DNA sequencing. For each gene, both antisense

and sense probes were synthesized with 1µg of linearized template by in vitro transcription

using DIG-RNA labelling kit (Roche) following the manufacturer's instruction. The probes

were ethanol precipitated and dissolved in 100 µL of hybridization buffer (50% formamide; 5

x standard saline citrate [SSC], pH 4.5; 50 µg/mL yeast tRNA; 1% sodium dodecyl sulfate

[SDS]; and 50 µg/mL heparin). The probes were stored at –80°C until use.

ACC

EPTE

D M

ANU

SCR

IPT



Tissue preparation. Liver sections were prepared as for laser microdissection described

above.

Hybridization. The slides were dried on a hot plate 5 min at 35°C prior to fixation in

paraformaldehyde 4% in PBS at 4°C for 10min followed by 3 washes in PBS for 3 minutes

each. Antisense or sense RNA probe was diluted to a final concentration of 1 ng/µL in

hybridization buffer, and incubated with the tissue in a humidified chamber at 70°C for 18

hours. Free probes were removed by a sequential washing: 3 times in SSC 1x, 50%

formamide, 0.1% Tween 20 for 30 min at 65°C and 2 times 30 min in 100mM maleic acid pH

7.5, 150mM NaCl, 0.1% Tween 20 (MABT) at room temperature. The tissue was then

incubated in the blocking solution (MABT + 2% blocking reagent (Roche) + 20% inactivated

goat serum) 1 hour at room temperature. The hybridized RNA probes were detected by anti-

digoxigenin alkaline phosphatase (1:2500 dilution; Roche) in alkaline-phosphatase staining

buffer (NTMT (100mM NaCl, 50mM MgCl, 100mM Tris pH9.5, 0.1%Tween20 + NBT

(Promega) and BCIP (Promega)).

Cell ploidy

One million freshly isolated hepatocytes were gently permeabilized in PBS 0.5% saponin,

treated by 100µg/mL RNAse and stained with 10µg/mL propidium iodide. DNA content

analysis was performed on FACS Calibur cytometer (BD Biosciences) on the linear scaled

FL2-A. Ten thousand events were recorded by Cell Quest software (BD Biosciences), and

ploidy was calculated using Modfit 2.0 software (Verity Software House), after removing of

doublets on FL2-A versus FL2-W dot plots. Mouse lymphocytes were used as control for

diploidy.

ACC

EPTE

D M

ANU

SCR

IPT



Isolation of 4N- and 8N-enriched hepatocytes by flow cytometer

Freshly isolated hepatocytes were analyzed on FACS Calibur cytometer (BD Biosciences)

equipped with a catcher tube. Linear scaled SSC-H versus FSC-H dot plots were used to

define gates of the sort at low speed, in exclusion mode. Samples were collected, immediately

concentrated, and used for viability count, ploidy verification, cell culture, or RNA extraction.

Viability assay

Viability of sorted hepatocytes was assayed with 0.05% trypan-blue exclusion immediately

after cell sort, and by lactate dehydrogenase (LDH) release assay (LDH assay, Roche) on cell

culture [30] 24h post-seeding.

RNA extraction and quantitative real-time RT-PCR

Total RNAs (except for LCM samples) were extracted using SV total RNA isolation system

(Promega) according to manufacturer’s instructions. Quality and quantity of total RNA were

assayed on a lab chip device (Agilent 2100 Bioanalyser) or Nanodrop (Agilent). Two

micrograms of total RNAs were reverse-transcribed using random primers and MMLV

Reverse Transcriptase (Promega). Quantitative PCR was performed using qPCR MasterMix

Plus for SYBR green (Eurogentec, Seraing, Belgium) on ABI prism 7000 SDS (PE-

Biosystems): 95°C for 10 min and 40 cycles of 95°C for 15 seconds and 60°C for 1 min. Each

amplification was duplicated. We verified genomic DNA contamination by negative control

of reverse transcription and amplification efficiency by standard curves. Each result was

normalized with beta-actin endogenous value. Primers are presented in Table 1.

Statistical analysis

ACC

EPTE

D M

ANU

SCR

IPT



A p-value, from non parametric Mann Whitney or Spearman correlation tests (StatView

software), lower than 0.05 was considered as statistically significant.

ACC

EPTE

D M

ANU

SCR

IPT



RESULTS

Efficiency of differential zonal liver perfusion.

Selective destruction of periportal or perivenous areas of liver lobules by digitonin (Figure 1,

panels A, B and C) prior to tissue dissociation showed a classical reticular aspect of liver as

described in rat [29] and mouse livers [31]. The pellets consisted mainly of hepatocytes and

about 20x106 hepatocytes were collected per liver with cell viability above 80%. The

selectivity of liver destruction was evaluated by quantitative RT-PCR of intracellular

transcriptional level of Pepck (phosphoenolpyruvate carboxykinase) involved in glycogenesis,

known as a periportal mRNA marker [32, 33], and of Glutamine synthetase (Gs) (glutamine

synthesis) [33, 34] and Cyp2e1 (xenobiotic metabolism), as centrilobular mRNA markers [33-

36] (Figure 2). Pepck mRNAs were found to be mainly expressed in periportal areas (p<0.05),

whereas both Gs and Cyp2e1 mRNAs were mainly expressed in centrilobular areas (p<0.05)

(Figure 2A), validating the method.

Efficiency of laser capture microdissection.

Laser Capture Microdissection (LCM) was selected as a complementary approach allowing to

study various hepatic cell types, and not only hepatocytes. The histogen staining and

histological architecture allowed us to easily identify portal space and centrolobular vein

(Figure 3, panels A and D) with high confidence. No contamination between periportal and

perivenous areas (Figure 3, panels B and C) was observed while capturing the samples. As

expected (Figure 4A), we retrieved the upregulation of Pepck in periportal areas (p<0.05)

whereas Gs and Cyp2e1 mRNAs were predominant in perivenous areas (p<0.05). We noticed

that the amplitude of the mRNA level was higher with LCM than with the digitonin method.

ACC

EPTE

D M

ANU

SCR

IPT



Ceruloplasmin is overexpressed in periportal area according to the digitonin method.

We studied hepatic mRNA expression levels of 8 genes involved in iron metabolism in

periportal or perivenous hepatocytes. Results are shown in Figure 2B. Ceruloplasmin mRNA

levels were higher in periportal areas (p<0.05) compared to total-liver hepatocytes.

Furthermore, we also found that Tfr2 mRNA was overexpressed in periportal and perivenous

(p<0.05) cells compared to total-liver hepatocytes. No significant zonation was observed for

mRNA levels of Hepc1, Hjv, Ferroportin, Hfe, Tfr1, or L-Ferritin.

Ceruloplasmin and Tfr1 are overexpressed in periportal areas according to LCM.

We found that Ceruloplasmin and Tfr1 mRNA levels were higher in periportal areas than in

total liver (p<0.05) and versus perivenous areas (p<0.05) (Figure 4B). This double

significance confirmed the results. Hepc1, Hjv, ferroportin, Hfe, Tfr2, and L-ferritin mRNAs

were equally expressed along the portocentral axis of the lobules.

Ceruloplasmin is overexpressed in periportal areas according to in situ hybridization.

In order to confirm results obtained by LCM, we performed in situ hybridization of

ceruloplasmin and ferroportin mRNA with both antisense and sense probes (Figure 5 , panels

A and B, respectively). Ceruloplasmin expression was observed only in hepatocytes. No

detectable signal was observed using the ceruloplasmin sense probe (negative probe). The

ceruloplasmin staining was stronger in periportal hepatocytes (Figure 5A). The expression of

ferroportin was not uniform but did however not present any clear gradient. Ferroportin

staining was weakly detected in hepatocytes and better detected in sinusoidal cells, i.e Kupffer

cells (Figure 5B). No staining was detected with the ferroportin sens probe (negative probe).

ACC

EPTE

D M

ANU

SCR

IPT



These results are in agreement with our quantitative RT-PCR findings on LCM samples.

Furthermore, they confirmed the major implication of other hepatic cell types than

hepatocytes, namely Kupffer cells, of the ferroportin.

Relationship between hepatocyte ploidy and location within the lobule.

In this study, we isolated hepatocytes from periportal and perivenous areas by the digitonin-

collagenase protocol, and we directly measured cellular DNA content in these cells by flow

cytometry using propidium iodide staining (Table 2). As expected in adult mice, close to 97%

of hepatocytes were polyploid (ie >2N) [30, 37-39] (Table 2). We identified a depletion in

diploid cells in the centrilobular hepatocyte subpopulation (2.9%) compared to periportal cells

(6.6%) (p<0.001), and an enrichment in diploid hepatocytes in periportal areas (p<0.01)

compared to controls, demonstrating that diploid hepatocytes were preferentially located

around the portal triad. Interestingly, we also demonstrated that the different populations of

polyploid (ie 4N, 8N) hepatocytes were homogeneously distributed within the lobule.

Efficient enrichment of hepatocyte ploidy sort.

In order to isolate viable hepatocytes on the basis of DNA content, we developed a protocol to

sort, by flow cytometry, fresh polyploid hepatocytes from adult mouse hepatocytes on SSC-H

(granularity/cytoplasmic complexity) versus FSC-H (size) parameters on linear scales without

using dye (Figure 5A). Sorted subpopulations were homogeneous in ploidy (Figure 5B), size,

and granularity (Figure 5A). Enrichment of the 4N subpopulation was up to 96.8%, and

72.7% for the 8N hepatocytes (Figure 5C). Because diploid cells represented less than 4% of

total hepatocytes, they were not sorted from adult mice liver. Moreover, propidium iodide

staining of these subpopulations revealed heterogeneous populations in terms of nuclearity,

with mono- and bi-nucleated cells (Figure 5D). Immediately after the sorting, as well as 24h

ACC

EPTE

D M

ANU

SCR

IPT



after seeding, viability was slightly higher in sorted subpopulations than in non-sorted whole

population maintained under the same conditions, as evaluated by trypan-blue dye exclusion

and by LDH release respectively (Table 3). High quality total RNA was obtained with a

28S/18S ratio of more than 1.7 as measured on a lab chip Agilent device. Moreover, cells

were kept in culture up to 72h, demonstrating the viability and sterility of this protocol (data

not shown).

Gene expression and hepatocyte ploidy.

We assayed transcriptional levels of periportal (Pepck) and centrilobular (Gs and Cyp2e1)

markers in 4N- and 8N-hepatocytes (Figure 6A). Pepck was uniformly expressed in 4N-

enriched subpopulation, 8N-enriched subpopulation and total hepatocyte population. Gs

mRNAs were overexpressed in 4N-enriched versus 8N-enriched and total-liver hepatocyte

populations (p<0.05). Cyp2e1 mRNAs were also overexpressed in the 4N-enriched

subpopulation compared to 8N-enriched subpopulation and total hepatocyte population

(p<0.05). In these samples, we observed a positive correlation between 18S rRNA and Beta-

Actin mRNA (rho=0.676, p<0.05) (Figure 6B) suggesting that global gene expression in

hepatocytes was effectively related to DNA content unit. Previous data indicated that

hepatocyte cell ploidy was correlated to nuclear RNA synthesis, RNA polymerase activity,

and cellular RNA content [40, 41].

mRNA levels of genes involved in iron metabolism, Hepc1, Hjv, ferroportin, Hfe, Tfr1, Tfr2

and L-ferritin, did not vary significantly between 4N-enriched, 8N-enriched subpopulations

and total hepatocyte population (Figure 6C). However, we only observed a trend towards a

decrease in expression of ceruloplasmin in these polyploid cells compared to total liver

hepatocytes suggesting that diploid hepatocytes are the main source of this protein.

ACC

EPTE

D M

ANU

SCR

IPT



DISCUSSION

In this study, we addressed the question of a zonation of iron metabolism mRNA expression

in basal condition which could secondarily play a role during iron overload diseases.

Periportal or centrilobular hepatic cells were collected either by specific zonation destruction

with digitonin prior to liver dissociation or by laser capture microdissection (LCM). Liver

dissociation and LCM allowed us to obtain gene expression results in hepatocytes only and in

hepatic cells, including sinusoidal cells, respectively. Furthermore, these methods allowed us

to measure mRNA levels by quantitative PCR, a more quantitative methodology than mRNA

in situ hybridization.

As expected, overexpression of Pepck mRNAs [32, 33] in periportal areas and overexpression

of glutamine synthetase [33, 34] and Cyp2e1 [35, 36] mRNAs in centrilobular areas validated

the efficiency of both methods. However, differences revealed by LCM were sharper between

periportal and perivenous areas. This was well illustrated by glutamine synthetase, an

hepatocyte specific gene (Figures 2 and 4). A limitation of the digitonin-dissociation

methodology is that periportal, perivenous and control cells were not originating from the

same animals, a potential source of increased variability. By contrast with digitonin-

dissociation methodology, LCM-extracted areas contained a majority of hepatocytes but also

non parenchymal cells. This feature could explain our data on mRNA level of Tfr1, which is

overexpressed in periportal areas of adult mice using LCM method only. This suggests that

non hepatocyte cell types could be involved in the portocentral decreasing gradient of Tfr1, in

accordance with data reporting changes in Tfr1 protein expression in aging rats [42]. Indeed,

Tfr1 protein staining was mainly observed within sinusoidal cells and in periportal areas in

adult rats, whereas it was parenchymal and perivenous in younger rats.

ACC

EPTE

D M

ANU

SCR

IPT



Using mRNA in situ hybridization, previous data reported that Hfe, Tfr2, Tf, Tfr1 and L-

ferritin mRNAs in rat [12, 13] and hepcidin mRNA in mouse [14] were homogeneously

expressed within the lobules in the hepatocytes. We found accordingly that, in the liver of

adult C57BL/6 mouse, hepcidin 1, Hfe, Tfr2 and L-ferritin mRNAs did not show a

portocentrolobular gradient. However, we found that isolated hepatocytes expressed Tfr2 with

an unusual pattern corresponding to both periportal and centrilobular veins. A better

understanding of the exact role of TfR2 gene, which is not directly implicated in hepatocyte

iron uptake, in the control of hepcidin expression [20] and more globally in iron metabolism

will help to understand such pattern.

Haemojuvelin is a major regulator of hepcidin expression [18] which is a negative regulator

of the iron export activity of ferroportin [43]. We did not detect any zonation for either

ferroportin or hemojuvelin mRNAs at variance with authors who found that the levels of

ferroportin mRNA in the rat [12] and activity of hemojuvelin promoter in the mouse [14]

were higher in periportal hepatocytes. We confirmed our data on ferroportin expression by in

situ hybridization. A species-difference could be possible for ferroportin expression. The

difference with previous published results for hemojuvelin could suggest : i) a strain

specificity [44-46], ii) a perturbation in the hemojuvelin promoter activity or, iii) an alteration

of the upstream-transcript stability in periportal areas in hemojuvelin-deficient transgenic

mice [14].

Using the ploidy approach, we found that the level of hepcidin 1, hemojuvelin, ferroportin,

Hfe, Tfr1, Tfr2, ceruloplasmin and L-ferritin mRNAs normalized on actin mRNA level were

similar in 4N- and 8N-enriched hepatocyte subpopulations compared to total-liver

hepatocytes suggesting that the transcriptional activity of theses genes is correlated with cell

ACC

EPTE

D M

ANU

SCR

IPT



DNA content. Several authors have explored the relationship between ploidy and functional

activity. Some have found that the rate of protein synthesis was directly proportional to the

degree of cell ploidy [47-52], others have demonstrated that rat polyploid-enriched fractions

showed higher cytochrome P450 [5], and glutamine synthetase [6] activities and, in contrast,

that diploid hepatocytes presented a higher ceruloplasmin biosynthetic rate [5], suggesting

exogenous control factors of these gene expressions, such as cell-cell contact or circulating

soluble factors. Our results on Cyp2e1, Gs and ceruloplasmin mRNAs expression reinforce

this view.

Our data, indicating about a two-fold induction of ceruloplasmin mRNA level in periportal

areas, could be associated to the periportal increase of ceruloplasmin protein synthesis

previously reported in rat [5]. We also confirmed the increase in ceruloplasmin mRNA

expression in periportal hepatocytes by in situ hybridization. Ceruloplasmin is a ferroxidase,

which is secreted in the plasma by the liver. Its function depends on the presence of an iron

exporter, such as ferroportin, a membrane protein. The presence of extracellular

ceruloplasmin is required for proper iron export. Thus, ceruloplasmin deficiency had been

previously associated with the development of iron overload in mice [53] and in humans,

since patients with aceruloplasminemia present iron overload, including liver iron overload

[54]. Furthermore, at local level, high modulation of ceruloplasmin was reported in Usf2

knock-out mice, exhibiting iron overload secondary to the lack of hepcidin expression found

in this model [55]. Our results clearly showed that the expression patterns of ferroportin and

ceruloplasmin did not merge, suggesting that secretion site of ceruloplasmin differs from its

sites of action. However, we assume that the particular location of ceruloplasmin expression

in periportal hepatocytes could have a direct physiological implication in adjusting the level

of synthesis of Ceruloplasmin to serum iron. Ceruloplasmin could act as a tuner of iron

ACC

EPTE

D M

ANU

SCR

IPT



export, locally but also at systemic level. Moreover, the overexpression of ceruloplasmin in

periportal hepatocytes could explain, at least partially, the centrilobular accumulation of iron

in hepc1 knock out mice [11] by an extra iron efflux in periportal areas at the expense of a

weaker iron efflux in centrilobular areas. The role of periportal expression of ceruloplasmin in

disorders related to altered hepcidin-ferroportin pathways controlling iron export requires

further studies.

In conclusion, in this study, ceruloplasmin mRNA, and Tfr1 mRNA were the only iron-

metabolism genes which exhibited a decreasing mRNA level gradient along the portocentral

axis of hepatic lobules. Other genes, hepcidin 1, hemojuvelin, ferroportin, Hfe, and L-ferritin,

were not differentially expressed within liver lobules. Therefore, the hepatic iron distribution

observed during genetic haemochromatosis in liver is unlikely related to a static

transcriptional zonation. However, physiological implication of periportal location of

ceruloplasmin may be strategic in the control of iron metabolism, and should be taken into

account when considering the cell mechanisms involved in iron overload diseases

characterized by altered cellular iron export.

ACC

EPTE

D M

ANU

SCR

IPT



ACKNOWLEDGEMENTS

The authors thank Catherine Ribaud for technical assistance with mice care at INSERM U522

(Rennes), Pascale Bellaud for technical assistance on the laser microdissection facilities of

IFR140 (Rennes), Carole Gautier, Valérie Dupé and Audrey Fleury for helpful advises for in

situ hybridizations. This work was supported by INSERM, a PRIR nb.139 of the Région

Bretagne, the Association Fer et Foie, and the LSHM-CT-2006-037296 European Community

Grant.

ACC

EPTE

D M

ANU

SCR

IPT



REFERENCES

[1] K. Jungermann and T. Kietzmann, Zonation of parenchymal and nonparenchymal

metabolism in liver, Annu Rev Nutr 16 (1996) 179-203.

[2] R. Carriere, Polyploid cell reproduction in normal adult rat liver, Exp Cell Res 46

(1967) 533-40.

[3] W.Y. Brodsky and I.V. Uryvaeva, Cell polyploidy: its relation to tissue growth and

function, Int Rev Cytol 50 (1977) 275-332.

[4] S. Benhamouche, T. Decaens, C. Godard, R. Chambrey, D.S. Rickman, C. Moinard,

M. Vasseur-Cognet, C.J. Kuo, A. Kahn, C. Perret and S. Colnot, Apc tumor

suppressor gene is the "zonation-keeper" of mouse liver, Dev Cell 10 (2006) 759-70.

[5] P. Rajvanshi, D. Liu, M. Ott, S. Gagandeep, M.L. Schilsky and S. Gupta,

Fractionation of rat hepatocyte subpopulations with varying metabolic potential,

proliferative capacity, and retroviral gene transfer efficiency, Exp Cell Res 244 (1998)

405-19.

[6] J.C. Osypiw, R.L. Allen and D. Billington, Subpopulations of rat hepatocytes

separated by Percoll density-gradient centrifugation show characteristics consistent

with different acinar locations, Biochem J 304 ( Pt 2) (1994) 617-24.

[7] T.C. Iancu, Y. Deugnier, J.W. Halliday, L.W. Powell and P. Brissot, Ultrastructural

sequences during liver iron overload in genetic hemochromatosis, J Hepatol 27 (1997)

628-38.

[8] X.Y. Zhou, S. Tomatsu, R.E. Fleming, S. Parkkila, A. Waheed, J. Jiang, Y. Fei, E.M.

Brunt, D.A. Ruddy, C.E. Prass, R.C. Schatzman, R. O'Neill, R.S. Britton, B.R. Bacon

and W.S. Sly, HFE gene knockout produces mouse model of hereditary

hemochromatosis, Proc Natl Acad Sci U S A 95 (1998) 2492-7.

ACC

EPTE

D M

ANU

SCR

IPT



[9] C. Pigeon, B. Turlin, T.C. Iancu, P. Leroyer, J. Le Lan, Y. Deugnier, P. Brissot and O.

Loreal, Carbonyl-iron supplementation induces hepatocyte nuclear changes in

BALB/CJ male mice, J Hepatol 30 (1999) 926-34.

[10] P. Brissot, T.L. Wright, W.L. Ma and R.A. Weisiger, Efficient clearance of non-

transferrin-bound iron by rat liver. Implications for hepatic iron loading in iron

overload states, J Clin Invest 76 (1985) 1463-70.

[11] J.C. Lesbordes-Brion, L. Viatte, M. Bennoun, D.Q. Lou, G. Ramey, C. Houbron, G.

Hamard, A. Kahn and S. Vaulont, Targeted disruption of the hepcidin 1 gene results in

severe hemochromatosis, Blood 108 (2006) 1402-5.

[12] A.S. Zhang, S. Xiong, H. Tsukamoto and C.A. Enns, Localization of iron metabolism-

related mRNAs in rat liver indicate that HFE is expressed predominantly in

hepatocytes, Blood 103 (2004) 1509-14.

[13] K.A. Basclain and G.P. Jeffrey, Coincident increase in periportal expression of iron

proteins in the iron-loaded rat liver, J Gastroenterol Hepatol 14 (1999) 659-68.

[14] V. Niederkofler, R. Salie and S. Arber, Hemojuvelin is essential for dietary iron

sensing, and its mutation leads to severe iron overload, J Clin Invest 115 (2005) 2180-

6.

[15] C. Pigeon, G. Ilyin, B. Courselaud, P. Leroyer, B. Turlin, P. Brissot and O. Loreal, A

new mouse liver-specific gene, encoding a protein homologous to human

antimicrobial peptide hepcidin, is overexpressed during iron overload, J Biol Chem

276 (2001) 7811-9.

[16] G. Nicolas, M. Bennoun, I. Devaux, C. Beaumont, B. Grandchamp, A. Kahn and S.

Vaulont, Lack of hepcidin gene expression and severe tissue iron overload in upstream

stimulatory factor 2 (USF2) knockout mice, Proc Natl Acad Sci U S A 98 (2001)

8780-5.

ACC

EPTE

D M

ANU

SCR

IPT



[17] J.N. Feder, D.M. Penny, A. Irrinki, V.K. Lee, J.A. Lebron, N. Watson, Z. Tsuchihashi,

E. Sigal, P.J. Bjorkman and R.C. Schatzman, The hemochromatosis gene product

complexes with the transferrin receptor and lowers its affinity for ligand binding, Proc

Natl Acad Sci U S A 95 (1998) 1472-7.

[18] G. Papanikolaou, M.E. Samuels, E.H. Ludwig, M.L. MacDonald, P.L. Franchini, M.P.

Dube, L. Andres, J. MacFarlane, N. Sakellaropoulos, M. Politou, E. Nemeth, J.

Thompson, J.K. Risler, C. Zaborowska, R. Babakaiff, C.C. Radomski, T.D. Pape, O.

Davidas, J. Christakis, P. Brissot, G. Lockitch, T. Ganz, M.R. Hayden and Y.P.

Goldberg, Mutations in HFE2 cause iron overload in chromosome 1q-linked juvenile

hemochromatosis, Nat Genet 36 (2004) 77-82.

[19] H. Kawabata, R. Yang, T. Hirama, P.T. Vuong, S. Kawano, A.F. Gombart and H.P.

Koeffler, Molecular cloning of Transferrin Receptor 2. A new member of the

transferrin receptor-like family, J Biol Chem 274 (1999) 20826-20832.

[20] E. Nemeth, A. Roetto, G. Garozzo, T. Ganz and C. Camaschella, Hepcidin is

decreased in TFR2 hemochromatosis, Blood 105 (2005) 1803-6.

[21] P.A. Seligman, R.B. Schleicher and R.H. Allen, Isolation and characterization of the

transferrin receptor from human placenta, J Biol Chem 254 (1979) 9943-6.

[22] G.C. Ford, P.M. Harrison, D.W. Rice, J.M. Smith, A. Treffry, J.L. White and J. Yariv,

Ferritin: design and formation of an iron-storage molecule, Philos Trans R Soc Lond B

Biol Sci 304 (1984) 551-65.

[23] S. Levi, S.J. Yewdall, P.M. Harrison, P. Santambrogio, A. Cozzi, E. Rovida, A.

Albertini and P. Arosio, Evidence of H- and L-chains have co-operative roles in the

iron-uptake mechanism of human ferritin, Biochem J 288 (1992) 591-6.

[24] A.T. McKie, P. Marciani, A. Rolfs, K. Brennan, K. Wehr, D. Barrow, S. Miret, A.

Bomford, T.J. Peters, F. Farzaneh, M.A. Hediger, M.W. Hentze and R.J. Simpson, A

ACC

EPTE

D M

ANU

SCR

IPT



novel duodenal iron-regulated transporter, IREG1, implicated in the basolateral

transfer of iron to the circulation, Molecular Cell 5 (2000) 299-309.

[25] A. Donovan, A. Brownlie, Y. Zhou, J. Shepard, S.J. Pratt, J. Moynihan, B.H. Paw, A.

Drejer, B. Barut, A. Zapata, T.C. Law, C. Brugnara, S.E. Lux, G.S. Pinkus, J.L.

Pinkus, P.D. Kingsley, J. Palis, M.D. Fleming, N.C. Andrews and L.I. Zon, Positional

cloning of zebrafish ferroportin1 identifies a conserved vertebrate iron exporter,

Nature 403 (2000) 776-81.

[26] S. Abboud and D.J. Haile, A novel mammalian iron-regulated protein involved in

intracellular iron metabolism, J Biol Chem 275 (2000) 19906-12.

[27] C.K. Mukhopadhyay, Z.K. Attieh and P.L. Fox, Role of ceruloplasmin in cellular iron

uptake, Science 279 (1998) 714-7.

[28] K.O. Lindros and K.E. Penttila, Digitonin-collagenase perfusion for efficient

separation of periportal or perivenous hepatocytes, Biochem J 228 (1985) 757-60.

[29] E. Wodey, A. Fautrel, M. Rissel, M. Tanguy, A. Guillouzo and Y. Malledant,

Halothane-induced cytotoxicity to rat centrilobular hepatocytes in primary culture is

not increased under low oxygen concentration, Anesthesiology 79 (1993) 1296-303.

[30] M.B. Troadec, B. Courselaud, L. Detivaud, C. Haziza-Pigeon, P. Leroyer, P. Brissot

and O. Loreal, Iron overload promotes Cyclin D1 expression and alters cell cycle in

mouse hepatocytes, J Hepatol 44 (2006) 391-9.

[31] H. Taniai, I.N. Hines, S. Bharwani, R.E. Maloney, Y. Nimura, B. Gao, S.C. Flores,

J.M. McCord, M.B. Grisham and T.Y. Aw, Susceptibility of murine periportal

hepatocytes to hypoxia-reoxygenation: role for NO and Kupffer cell-derived oxidants,

Hepatology 39 (2004) 1544-52.

ACC

EPTE

D M

ANU

SCR

IPT



[32] J.M. Ruijter, R.G. Gieling, M.M. Markman, J. Hagoort and W.H. Lamers,

Stereological measurement of porto-central gradients in gene expression in mouse

liver, Hepatology 39 (2004) 343-52.

[33] V.M. Christoffels, H. Sassi, J.M. Ruijter, A.F. Moorman, T. Grange and W.H.

Lamers, A mechanistic model for the development and maintenance of portocentral

gradients in gene expression in the liver, Hepatology 29 (1999) 1180-92.

[34] A. Cadoret, C. Ovejero, B. Terris, E. Souil, L. Levy, W.H. Lamers, J. Kitajewski, A.

Kahn and C. Perret, New targets of beta-catenin signaling in the liver are involved in

the glutamine metabolism, Oncogene 21 (2002) 8293-301.

[35] G.T. Wagenaar, R.A. Chamuleau, J.G. de Haan, M.A. Maas, P.A. de Boer, F. Marx,

A.F. Moorman, W.M. Frederiks and W.H. Lamers, Experimental evidence that the

physiological position of the liver within the circulation is not a major determinant of

zonation of gene expression, Hepatology 18 (1993) 1144-53.

[36] L. Chen, G.J. Davis, D.W. Crabb and L. Lumeng, Intrasplenic transplantation of

isolated periportal and perivenous hepatocytes as a long-term system for study of

liver-specific gene expression, Hepatology 19 (1994) 989-98.

[37] P.O. Seglen, DNA ploidy and autophagic protein degradation as determinants of

hepatocellular growth and survival, Cell Biol Toxicol 13 (1997) 301-15.

[38] J.C. Garrison, T.U. Bisel, P. Peterson and E.M. Uyeki, Changes in hepatocyte ploidy

in response to chromium, analyzed by computer-assisted microscopy, Fundam Appl

Toxicol 14 (1990) 346-55.

[39] S. Madra, J. Styles and A.G. Smith, Perturbation of hepatocyte nuclear populations

induced by iron and polychlorinated biphenyls in C57BL/10ScSn mice during

carcinogenesis, Carcinogenesis 16 (1995) 719-27.

ACC

EPTE

D M

ANU

SCR

IPT



[40] I.R. Johnston, A.P. Mathias, F. Pennington and D. Ridge, Distribution of RNA

polymerase activity among the various classes of liver nuclei, Nature 220 (1968) 668-

72.

[41] P.O. Seglen, Protein-catabolic stage of isolated rat hepatocytes, Biochim Biophys Acta

496 (1977) 182-91.

[42] R. Sciot, G. Verhoeven, P. Van Eyken, J. Cailleau and V.J. Desmet, Transferrin

receptor expression in rat liver: immunohistochemical and biochemical analysis of the

effect of age and iron storage, Hepatology 11 (1990) 416-27.

[43] E. Nemeth, M.S. Tuttle, J. Powelson, M.B. Vaughn, A. Donovan, D.M. Ward, T.

Ganz and J. Kaplan, Hepcidin regulates cellular iron efflux by binding to ferroportin

and inducing its internalization, Science 306 (2004) 2090-3.

[44] R.C. Leboeuf, D. Tolson and J.W. Heinecke, Dissociation between tissue iron

concentrations and transferrin saturation among inbred mouse strains, J Lab Clin Med

126 (1995) 128-36 issn: 0022-2143.

[45] M. Bensaid, S. Fruchon, C. Mazeres, S. Bahram, M.P. Roth and H. Coppin,

Multigenic control of hepatic iron loading in a murine model of hemochromatosis,

Gastroenterology 126 (2004) 1400-8.

[46] B. Courselaud, M.B. Troadec, S. Fruchon, G. Ilyin, N. Borot, P. Leroyer, H. Coppin,

P. Brissot, M.P. Roth and O. Loreal, Strain and gender modulate hepatic hepcidin 1

and 2 mRNA expression in mice, Blood Cells Mol Dis 32 (2004) 283-9.

[47] N.C. Martin, C.T. McCullough, P.G. Bush, L. Sharp, A.C. Hall and D.J. Harrison,

Functional analysis of mouse hepatocytes differing in DNA content: volume, receptor

expression, and effect of IFNgamma, J Cell Physiol 191 (2002) 138-44.

ACC

EPTE

D M

ANU

SCR

IPT



[48] E. Le Rumeur, C. Beaumont, C. Guillouzo, M. Rissel, M. Bourel and A. Guillouzo,

All normal rat hepatocytes produce albumin at a rate related to their degree of ploidy,

Biochem Biophys Res Commun 101 (1981) 1038-46.

[49] A. Tulp, J. Welagen and P. Emmelot, Separation of intact rat hepatocytes and rat liver

nuclei into ploidy classes by velocity sedimentation at unit gravity, Biochim Biophys

Acta 451 (1976) 567-82.

[50] C.J. Van Noorden, I.M. Vogels, G. Fronik and R.D. Bhattacharya, Ploidy class-

dependent variations during 24 h of glucose-6-phosphate and succinate dehydrogenase

activity and single-stranded RNA content in isolated rat hepatocytes, Exp Cell Res 155

(1984) 381-8.

[51] C.J. Van Noorden, I.M. Vogels, J.M. Houtkooper, G. Fronik, J. Tas and J. James,

Glucose-6-phosphate dehydrogenase activity in individual rat hepatocytes of different

ploidy classes. I. Developments during postnatal growth, Eur J Cell Biol 33 (1984)

157-62.

[52] E. Le Rumeur, C. Guguen-Guillouzo, C. Beaumont, A. Saunier and A. Guillouzo,

Albumin secretion and protein synthesis by cultured diploid and tetraploid rat

hepatocytes separated by elutriation, Exp Cell Res 147 (1983) 247-54.

[53] Z.L. Harris, A.P. Durley, T.K. Man and J.D. Gitlin, Targeted gene disruption reveals

an essential role for ceruloplasmin in cellular iron efflux, Proc Natl Acad Sci U S A 96

(1999) 10812-7.

[54] N.E. Hellman, M. Schaefer, S. Gehrke, P. Stegen, W.J. Hoffman, J.D. Gitlin and W.

Stremmel, Hepatic iron overload in aceruloplasminaemia, Gut 47 (2000) 858-60.

[55] L. Viatte, J.C. Lesbordes-Brion, D.Q. Lou, M. Bennoun, G. Nicolas, A. Kahn, F.

Canonne-Hergaux and S. Vaulont, Deregulation of proteins involved in iron

metabolism in hepcidin-deficient mice, Blood 105 (2005) 4861-4.

ACC

EPTE

D M

ANU

SCR

IPT



FIGURE LEGENDS

Figure 1: Selective zonal destruction in liver mouse.

To obtain periportal or perivenous hepatocytes, mouse liver was perfused by inferior vena

cava or portal vein by digitonin, prior to classical liver dissociation. (A) Destruction of

selected zones showed a typical reticular aspect; white zones correspond to dead cells in

periportal (B) or perivenous areas (C). Original magnification x1.5 (A) and x10 (B and C).

Figure 2: mRNA expression levels in periportal and perivenous hepatocytes obtained by

liver perfusion.

mRNA levels were obtained by quantitative RT-PCR, and expressed as log2(zones/control).

Controls corresponded to total-liver hepatocytes. Mean +/- SD. *p<0.05, Mann Whitney test,

between samples and control (single star), or between centrilobular and periportal cells (star

with bracket). (A) mRNA levels of markers from periportal or perivenous liver zones. Pepck:

phosphoenolpyruvate carboxykinase, Gs: glutamine synthetase, Cyp2e1: cytochrome P450

2e1. (B) mRNA levels of genes implicated in iron metabolism. Hepcidin (Hepc1),

hemojuvelin (Hjv), ferroportin, ceruloplasmin, Hfe, transferrin receptor 1 (Tfr1), transferrin

receptor 2 (Tfr2) and L-ferritin.

Figure 3: Laser capture microdissection of liver lobules.

Thin 10 µm slides of liver were stained by histogen (panels A to C). (A1) Typical portal space

with portal vein (pv) and hepatic artery (ha) and biliary canaliculi (bc), and centrolobular vein

(cv) are well seen (original magnification x 200). From these structures, hepatocytes from

both periportal and perinous areas can be selected. (A2) Different areas were selected on the

ACC

EPTE

D M

ANU

SCR

IPT



basis of the histological features of periportal to perivenous areas and the histogen staining.

We captured first the periportal areas (panels B1 and B2), then the perivenous areas (panels

C1 and C2). B1 and C1 show the remaining tissue whereas B2 and C2 show the captured

samples from which RNAs were extracted. Original magnification x20.

Figure 4: mRNA expression levels in periportal and perivenous liver tissue isolated by

Laser Capture Microdissection.

mRNA levels were obtained by quantitative RT-PCR, and expressed as log2(captured

zones/captured total liver). Mean +/- SD. *p<0.05, Mann Whitney test, between captured

areas and captured total liver (single star), or between perivenous and periportal cells (star

with bracket). (A) mRNA levels of markers from periportal or perivenous liver zones. Pepck:

phosphoenolpyruvate carboxykinase, Gs: glutamine synthetase, Cyp2e1: cytochrome P450

2e1. (B) mRNA levels of genes implicated in iron metabolism. Hepcidin (Hepc1),

hemojuvelin (Hjv), ferroportin, ceruloplasmin, Hfe, transferrin receptor 1 (Tfr1), transferrin

receptor 2 (Tfr2) and L-ferritin.

Figure 5: In situ hybridization analysis of hepatic ceruloplasmin and ferroportin

mRNAs expression. In situ hybridization analysis of Ceruloplasmin (panel A) and

Ferroportin (panel B) in mouse liver. For each gene, the images from the analysis of both

antisense (gene-specific probe) and sense probes (negative control) are shown. Incubation

times for the development: ceruloplasmin: 50h ; Ferroportin: 60h. * indicates Kupffer cells.

Pv indicates portal vein, cv, centrilobular vein. Original magnification x200.

Figure 6: Separation of adult mouse hepatocytes on size and granularity criteria.

ACC

EPTE

D M

ANU

SCR

IPT



Viable mouse hepatocytes were sorted on (A) SSC-H (granularity/complexity) versus FSC-H

(size) parameters on flow cytometer without using any dye, detergent or fixation. (B) An

aliquot of cells was stained by propidium iodide and analysed for cell DNA content on FL2-A

by Cell Quest software. (C) Distribution of ploidy was analysed by Modfit software in 18-20-

week old C57BL/6 mice (n=8) and in sorted hepatocytes (n=3 each subpopulation). Mean +/-

SD of percentage of total cell population. (D) Nuclearity was visualized by a propidium

iodide staining and revealed an heterogeneity of 4N- and 8N-enriched subpopulations.

Figure 7: mRNA expression levels in 4N- and 8N-enriched hepatocyte subpopulations.

mRNA levels were obtained by quantitative RT-PCR, and expressed as

log2(subpopulation/control). Controls correspond to total-liver hepatocytes. Mean +/- SD.

*p<0.05, Mann Whitney test between samples and control (single star), or between 4N- and

8N-enriched hepatocytes (star with braket). (A) mRNA levels of markers from periportal or

perivenous liver zones. Pepck: phosphoenolpyruvate carboxykinase, Gs: Glutamine

synthetase, Cyp2e1: cytochrome p450 2E1. (B) Correlation of Cycle threshold values (Ct) of

18S rRNA and actin mRNA (n=12) *p<0.05, Spearman test. (C) mRNA levels of genes

implicated in iron metabolism. Hepcidin (Hepc1), Hemojuvelin (Hjv), Ferroportin,

Ceruloplasmin, Hfe, Transferrin receptor 1 (Tfr1), Transferrin receptor 2 (Tfr2) and L-

Ferritin.

ACC

EPTE

D M

ANU

SCR

IPT



Symbol Gene name Official gene

symbol

Forward Reverse

Pepck phosphoenolpyruvate

carboxykinase

Pck1 ccacagctgctgcagaaca gaagggtcgcatggcaaa

Cyp2e1 cytochrome P450 2E1 Cyp2e1 tgcagtccgagacaggatga ggacgaggttgatgaatctctga

Gs glutamine synthetase Glul caggctgccataccaacttca tcctcaatgcacttcagaccat

18S - tgcaattattccccatgaacg gcttatgacccgcacttactgg

Actin beta actin Actb gacggccaagtcatcactattg ccacaggattccatacccaaga

Hepc1 hepcidin1 Hamp cctatctccatcaacagatg aacagataccacactgggaa

Hjv hemojuvelin Hfe2 aagtgggcattgtctggcag gttggtgccagtctccaaaag

Ferroportin ferroportin Slc40a1 gctgctagaatcggtctttggt cagcaactgtgtcaccgtcaa

Ceruloplasmin ceruloplasmin Cp gggagccgtctaccctgataa ttgtcatcagcccgttgaaa

Hfe Hfe Hfe gagcaagtgtgccccctccaagtctt aaggaaggcttcaggaggaacc

Tfr1 transferrin receptor 1 Tfrc tcatgagggaaatcaatgatcgta gccccagaagatatgtcggaa

Tfr2 transferrin receptor 2 Tfr2 agtggcgacgtttggaaca tcaggcacctcctttgcc

L-Ferritin L-ferritin Ftl cagtctgcaccgtctcttcg gtcatggctgatccggagtag

Table 1: Primers used for quantitative real-time RT-PCR.

ACC

EPTE

D M

ANU

SCR

IPT



Cell ploidy (%)

hepatocyte location n= 2N 4N 8N

periportal 6 6.6 +/-1.4 (p<0.05/control) 65.0 +/-4.8 (ns) 28.4 +/-4.4 (ns)

centrilobular 8 2.9 +/-1.6 (p<0.001/periportal) 66.5 +/-5.2 (ns) 30.6 +/-4.8 (ns)

total-liver hepatocytes 2 3.0 (2.3-3.8) 58.7 (51.5-65.8) 38.3 (31.8-44.8)

Table 2: Hepatocyte ploidy in the hepatic periportal and perivenous areas.

Isolation of periportal or perivenous hepatocytes was performed by a selective destruction of

liver areas by digitonin prior to classical liver dissociation. Cell ploidy was evaluated on

propidium iodide staining of hepatocytes on flow cytometer. Mean+/- SD. Mann Whitney

test. ns: non significant

ACC

EPTE

D M

ANU

SCR

IPT



Hepatocyte subpopulation n= Viability (%) Cell death

(mU LDH release/mg proteins)

4N-enriched hepatocytes 3 86.7 +/- 6.1 4266.1 +/- 902 *

8N-enriched hepatocytes 3 94.2 +/- 2.0 * 4657.4 +/- 1002 *

Total liver cell population 3 76.3 +/- 5.5 6413.1 +/- 1333

Table 3: Viability of 4N- and 8N-enriched hepatocyte subpopulations.

4N- and 8N-enriched subpopulations were obtained from 20-week old C57BL/6 mouse

hepatocytes. Immediately after sorts, hepatocyte viability was assayed by trypan-blue

exclusion. Sorted hepatocytes were plated in culture and cell death was evaluated by LDH

release assay (mU/mg proteins), 24h after seeding. Mean+/-SD. *p<0.05, Mann Whitney test.

ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



ACC

EPTE

D M

ANU

SCR

IPT



Documents

Transcripts of ceruloplasmin but not hepcidin, both major